CN117786941A - Industrial virtual-real bidirectional interaction system and method - Google Patents
Industrial virtual-real bidirectional interaction system and method Download PDFInfo
- Publication number
- CN117786941A CN117786941A CN202311531199.6A CN202311531199A CN117786941A CN 117786941 A CN117786941 A CN 117786941A CN 202311531199 A CN202311531199 A CN 202311531199A CN 117786941 A CN117786941 A CN 117786941A
- Authority
- CN
- China
- Prior art keywords
- entity
- simulation
- operation result
- sensor
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000003993 interaction Effects 0.000 title claims abstract description 35
- 230000002457 bidirectional effect Effects 0.000 title claims abstract description 32
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000004088 simulation Methods 0.000 claims abstract description 154
- 238000012360 testing method Methods 0.000 claims abstract description 72
- 230000002159 abnormal effect Effects 0.000 claims abstract description 57
- 238000012937 correction Methods 0.000 claims abstract description 33
- 238000001514 detection method Methods 0.000 claims description 30
- 230000002452 interceptive effect Effects 0.000 claims description 18
- 238000012544 monitoring process Methods 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 13
- 238000012216 screening Methods 0.000 claims description 13
- 238000004891 communication Methods 0.000 claims description 10
- 238000004458 analytical method Methods 0.000 claims description 9
- 230000005540 biological transmission Effects 0.000 claims description 9
- 238000011156 evaluation Methods 0.000 claims description 9
- 230000015556 catabolic process Effects 0.000 claims description 8
- 238000006731 degradation reaction Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 7
- 230000000295 complement effect Effects 0.000 claims description 3
- 238000013450 outlier detection Methods 0.000 claims description 2
- 230000008859 change Effects 0.000 abstract description 8
- 238000010586 diagram Methods 0.000 description 17
- 230000015654 memory Effects 0.000 description 8
- 239000011159 matrix material Substances 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 238000003745 diagnosis Methods 0.000 description 4
- 230000004927 fusion Effects 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000005856 abnormality Effects 0.000 description 3
- 230000036541 health Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 230000003595 spectral effect Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000003862 health status Effects 0.000 description 2
- 230000007787 long-term memory Effects 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 230000006403 short-term memory Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention provides an industrial virtual-real bidirectional interaction system and method, wherein the system comprises: the system comprises an entity system, a simulation system and a correction module; the entity system is used for collecting test data and transmitting the test data to the correction module; the correction module is used for outputting a parameter identification result of the entity system in the current system state based on the test data and updating the simulation system based on the parameter identification result; the test data is collected based on the physical system under any system state, wherein the system state is determined by at least one of working condition, service life and working environment of the physical system. The system and the method provided by the invention can distinguish the abnormal condition of the entity system from the condition, the service life period and the change of the working environment, and improve the practicability of data, thereby improving the system performance of the industrial virtual-real bidirectional interaction system.
Description
Technical Field
The invention relates to the technical field of information, in particular to an industrial virtual-real bidirectional interaction system and method.
Background
At present, as the application of the information physical fusion scheme in various fields is increasingly deep, the application of the information physical fusion scheme in the industrial field is increasingly important, and the transformation development of an industrial system can be driven. At present, the information physical fusion scheme mainly realizes information physical fusion by constructing an entity system and a virtual system (simulation system) corresponding to the entity system.
However, in the link of controlling the virtual system by the physical system, after the simulation model is built at a specific working condition point, the model parameters of the simulation model are rarely changed, so that the virtual system is not suitable for the requirements of generating virtual data and subsequent accompanying operation, control rate optimization and fault diagnosis of the dynamic system under the full working condition, full life period and each working environment.
Disclosure of Invention
The invention provides an industrial virtual-real bidirectional interaction system and method, which are used for solving the demand defect that a virtual system is not suitable for generating virtual data and subsequent accompanying operation, control rate optimizing and fault diagnosis of a dynamic system under all working conditions, full life time and working environments in the prior art.
The invention provides an industrial virtual-real bidirectional interaction system, which comprises: the system comprises an entity system, a simulation system and a correction module;
the entity system is used for collecting test data and transmitting the test data to the correction module;
the correction module is used for outputting a parameter identification result of the entity system in the current system state based on the test data and updating the simulation system based on the parameter identification result;
the test data is collected based on the physical system under any system state, wherein the system state is determined by at least one of working condition, service life and working environment of the physical system.
The invention provides an industrial virtual-real bidirectional interaction system, which also comprises an entity system monitoring module, wherein the entity system monitoring module is respectively in communication connection with the entity system and the simulation system;
the entity system monitoring module is used for respectively outputting an entity operation result and a simulation operation result based on the entity system and the simulation system under the same entity data, and carrying out error analysis on the entity operation result and the simulation operation result to judge the operation state of the entity system;
and continuously acquiring the entity operation result and the simulation operation result under the condition that the operation state is normal.
According to the industrial virtual-real bidirectional interaction system provided by the invention, the entity system monitoring module further comprises a fault detection module, wherein the fault detection module is used for detecting faults of the entity system under the condition that the running state is abnormal;
the fault detection module comprises a sensor evaluation unit, a sensor association network construction unit and a sensor abnormal point screening unit;
the sensor evaluation unit is used for acquiring the correlation among the sensors contained in the entity system;
the sensor association network construction unit is used for constructing a sensor association network of the entity system by taking the sensors as nodes and the correlation among the sensors as edges;
the sensor abnormal point screening unit is used for screening nodes of the sensor association network to obtain an abnormal sensor in the sensor association network.
According to the industrial virtual-real bidirectional interaction system provided by the invention, the sensor abnormal point screening unit further comprises an abnormal signal detection unit;
the abnormal signal detection unit is used for obtaining a fault detection result based on the transmission result of the abnormal signal corresponding to the abnormal node in the sensor association network.
According to the invention, the simulation system is used for controlling the entity system and the simulation system to run synchronously based on simulation control parameters, and adjusting the simulation control parameters and the control rate based on switching conditions of the entity system under different normal working conditions and different degradation states.
According to the invention, the simulation system is also used for acquiring a target simulation operation result output under a target working condition;
the target operating condition includes at least one of an extreme operating condition and an accident operating condition.
The invention provides an industrial virtual-real bidirectional interaction system, which also comprises a data processing unit;
and the data processing unit is used for carrying out missing data complement on the entity operation result, the simulation operation result and the test data.
The invention also provides an industrial virtual-real bidirectional interaction method, which comprises the following steps:
acquiring test data, wherein the test data is acquired based on an entity system in any system state;
based on the test data, obtaining a parameter identification result of the entity system in the current state;
updating a simulation system based on the parameter identification result;
the system state is determined by at least one of a working condition, a life span, and a working environment of the physical system.
According to the industrial virtual-real bidirectional interaction method provided by the invention, the simulation system is updated based on the parameter identification result, and then the method comprises the following steps:
acquiring an entity operation result and a simulation operation result which are respectively output by the entity system and the simulation system under the same physical data;
based on the entity operation result and the simulation operation result, performing error analysis on the entity operation result and the simulation operation result, and judging the operation state of the entity system;
and continuously acquiring the entity operation result and the simulation operation result under the condition that the operation state is normal.
According to the industrial virtual-real bidirectional interaction method provided by the invention, the operation state of the entity system is judged, and then the method further comprises the following steps:
under the condition that the running state is abnormal, acquiring the correlation among the sensors of the entity system, taking the sensors as nodes, and constructing a sensor correlation network of the entity system by taking the correlation surrounding edges among the sensors;
based on a sensor association network, obtaining an abnormal node in the sensor association network;
and obtaining a fault detection result corresponding to the abnormal node based on a transmission result of the abnormal signal corresponding to the abnormal node in the sensor association network.
According to the industrial virtual-real bidirectional interaction system and the method, the simulation system is updated through the test data collected by the entity system, so that the dynamic update of the simulation system is realized when the working condition, the service life period and the working environment change, the conditions of the abnormality of the entity system and the change of the working condition, the service life period and the working environment can be distinguished, and the system performance of the industrial virtual-real bidirectional interaction system is improved.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of an industrial virtual-real two-way interactive system according to the present invention;
FIG. 2 is a schematic diagram of a correction module according to the present invention;
FIG. 3 is a schematic diagram of a collarband matrix of a sensor association network provided by the present invention;
fig. 4 is a schematic diagram of a topology structure of a sensor association network provided by the present invention;
FIG. 5 is a block diagram of a long and short term memory network provided by the present invention;
FIG. 6 is a schematic diagram of a second embodiment of an industrial virtual-real two-way interactive system according to the present invention;
FIG. 7 is a schematic flow chart of the industrial virtual-real two-way interaction method provided by the invention;
FIG. 8 is a functional schematic of a communication interface provided by the present invention;
FIG. 9 is a schematic diagram of a connection between a simulation system and a correction module according to the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the industrial field, modeling errors inevitably exist between a simulation system and a physical system (a test bench) of the test bench, and the error accumulation effect of subsystems of different simulation systems can make the simulation system describe the physical system not well.
Aiming at the problems, the invention provides an industrial virtual-real bidirectional interactive system to realize dynamic adjustment of a simulation system. FIG. 1 is a schematic diagram of an industrial virtual-real two-way interactive system according to the present invention, as shown in FIG. 1, the system includes a physical system 110, a simulation system 120, and a correction module 130;
the entity system 110 is configured to collect test data and transmit the test data to the correction module 130;
the correction module 130 is configured to output a parameter identification result of the entity system 110 in a current system state based on the test data, and update the simulation system 120 based on the parameter identification result;
the test data is based on the physical system 110 being collected at any system state determined by at least one of the operating conditions, life span, operating environment of the physical system 110.
It should be noted that, in the industrial virtual-real bidirectional interaction system, information interaction between the systems may be implemented through communication interfaces among the entity system 110, the simulation system 120 and the modification module 130. Wherein the simulation system 120 may be connected in series or in parallel with the correction module 130. The communication interface herein may provide an interaction environment for interactions between the entity system 110 and the simulation system 120 as an intermediate for data exchange between the simulation system 120 and the entity system 110 of the controlled object. For example, system execution signals and process data of entity system 110 are collected and converted into digital signals required by simulation system 120; the data packets sent by the simulation system 120 may also be converted into various physical signals that can be collected by the gantry controller of the physical system 110.
Specifically, the actual data output as the test data may be performed by the entity system 110 in any system state for industrial operation. The system state herein may be determined by at least one of a certain operating condition of the entity system 110, a lifetime of the entity system 110, and an operating environment of the entity system 110. For example, the actual data output from the physical bench at the middle and late life spans can be used as test data. The test data herein may include numerical data, spectral data, and the like.
It should be noted that, in the actual application process of the entity system 110, before dynamically adjusting the simulation system 120, the current system state of the entity system 110 needs to be clarified, that is, whether the simulation system 120 needs to be updated by judging whether the current system state of the entity system 110 needs to be updated, for example, whether the simulation system 120 needs to be updated is achieved by setting the adjustment switch module, so that the adjusted simulation system 120 is more in line with the entity system 110 in the current system state. Thus, the test data may also include switching data, i.e., whether or not to update the simulation system 120 corresponding to the physical system 110 in the current system state. The switching data may be obtained by determining key operation parameters of the entity system 110 for reflecting the working condition, or may be determined by a value of a power value of a main pump of the entity system 110 when the working condition changes, for example, when the power value is smaller than a preset threshold, the switching data may be turned on, that is, the simulation system 120 for indicating the current system state is updated; the switch data may also be determined for the life span of the physical system 110, for example, a feature that reflects the health status of the physical system 110 is selected to construct a "health factor," which may be normalized to (0, 1) and marked with 0.1 as a significant degradation.
Here, the correction module 130 may be derived based on a recurrent neural network. Fig. 2 is a schematic structural diagram of a correction module provided in the present invention, and as shown in fig. 2, the correction module 130 may include an input layer, a hidden layer, and an output layer. Specifically, it may be determined whether system parameter identification is required for the test data based on the switching data in the test data after the correction module 130 receives the test data. When the switch data is off, the correction module 130 does not output the parameter identification result of the entity system 110; when the switching data is on, the correction module 130 may input the spectrum data, the numerical data, and the switching data in the test data into the input layer, and extract the data features of the test data in the time sequence and other dimensions through the input layer. And then, inputting the data characteristics into the hidden layer, analyzing and calculating system parameters corresponding to each test data through the hidden layer, and outputting a parameter identification result through the output layer. It should be noted that, the parameter identification result herein may reflect a gap between the system parameter of the entity system 110 in the current system state and the system parameter of the original entity system 110. Thus, the parameter identification result may be a system parameter value of the entity system 110 in the current system state, or may be a difference value of a system parameter of the entity system 110 between the current system state and the initial system state. Here, in the training stage of the correction module 130, nonlinear transformation may be performed through the input test data, and iterative training is performed on the correction module by depending on the actual parameter identification result output by the output layer and the labeling information of the test data, so as to find a better feature expression, and finally obtain the correction module capable of outputting an accurate parameter identification result.
Finally, the system parameters of the simulation system can be updated through the parameter identification result, so that the simulation system can realize the concomitant operation under various working conditions, service life and working environments.
It can be understood that, in the industrial virtual-real bidirectional interactive system, if the correction module 130 is not added, there may be a change in the working condition, the lifetime limit, and the working environment, and the simulation system 120 does not update along with the working condition, the lifetime limit, and the working environment, so that an error between the simulation data output by the simulation system 120 and the actual data output by the entity system 110 is larger, which results in a false alarm to the entity system. According to the industrial virtual-real bidirectional interactive system provided by the embodiment of the invention, the simulation system is updated based on the test data acquired by the entity system, so that the dynamic update of the simulation system is realized when the working condition, the service life period and the working environment change, the conditions of the abnormality of the entity system and the working condition, the service life period and the working environment change can be distinguished, and the system performance of the industrial virtual-real bidirectional interactive system is improved.
Based on any of the above embodiments, the industrial virtual-real two-way interaction system further includes an entity system monitoring module, where the entity system monitoring module establishes communication connection with the entity system 110 and the simulation system 120 respectively;
the entity system monitoring module is configured to obtain an operation state of the entity system 110 based on an entity operation result and a simulation operation result that are respectively output by the entity system 110 and the simulation system 120 under the same entity data;
and continuously acquiring the entity operation result and the simulation operation result under the condition that the operation state is normal.
It should be noted that, the entity system monitoring module may be communicatively connected to the entity system 110 and the simulation system 120, and the entity system detection module receives the entity operation result output by the entity system 110 and receives the simulation operation result output by the simulation system 120, so as to monitor the operation state of the entity system 110.
Specifically, the physical data of the physical system 110 may be transmitted to the simulation system 120, so that the simulation system 120 operates with the physical data. The physical data herein refers to system parameters such as control parameters including valve opening of a test bed, pump rotation speed, and the like. By acquiring the entity operation result and the simulation operation result respectively output by the entity system 110 and the simulation system 120 under the same physical data. And analyzing errors between the entity operation result and the simulation operation result. Whether the operation state of the physical system 110 is normal can be determined by whether the error converges.
It will be appreciated that the simulation system 120 updated by the modification module 130 may be considered a virtual system that is identical in system functionality and performance to the physical system 110. Thus, the simulation operation result output by the simulation system 120 under the same control parameter (physical data) may represent the physical operation result output by the physical system 110 with a normal operation state. Thus, when the error is converged, it indicates that the consistency between the simulation operation result output by the simulation system 120 and the entity operation result output by the entity system 110 is higher, and the operation state of the entity system is a normal state; when the error is not converged, it indicates that the consistency between the simulation operation result output by the simulation system 120 and the entity operation result output by the entity system 110 is low, and the operation state of the entity system is abnormal. It should be noted that, when the operation state of the entity system 110 is a normal state, the entity operation result and the simulation operation result may be continuously obtained, so as to realize real-time monitoring of the operation state of the entity system 110.
Based on any one of the above embodiments, the entity system monitoring module further includes a fault detection module, where the fault detection module is configured to perform fault detection on the entity system when the running state is abnormal;
the fault detection module comprises a sensor evaluation unit, a sensor association network construction unit and a sensor abnormal point screening unit;
the sensor evaluation unit is used for acquiring the correlation among the sensors contained in the entity system;
the sensor association network construction unit is used for constructing a sensor association network of the entity system by taking the sensors as nodes and the correlation among the sensors as edges;
the sensor abnormal point screening unit is used for screening nodes of the sensor association network to obtain an abnormal sensor in the sensor association network.
Specifically, in the entity system 110, firstly, due to the complex system structure and high signal coupling degree, a plurality of sensors in the entity system 110 form a sensor information network through the material flow, the energy flow and the signal flow of the entity system 110. Thus, the correlation between the sensors can be obtained by evaluating the correlation of the sensors by the sensor evaluation unit. For example, the maximum information correlation coefficient (MIC, the maximal information coefficient) can be used as a correlation evaluation index. Correlation herein refers to the degree of correlation of the sensor in terms of behavior and may include both statistically and temporally sequential behavior. For example, the correlation between the sensors can be calculated by the following formula:
where MIC is defined as the normalized mutual information of the partitions that enable the highest resolution of the sample points in all two-dimensional grids of two sensor variables, while MIC represents the maximum value in the matrix, i.e. the correlation between the two sensors. X represents a set of events for one sensor; y represents the event set of another sensor. Here, mutual information is a method for evaluating correlation in information theory, and measures correlation between two event sets. Wherein the average mutual information may be defined by the following formula:
mutual information may also be defined by the following formula:
where P represents the probability. Mutual information quantity I [ X ]; y ] in the joint probability space P (X, Y). The information I (X; Y) overcomes the randomness of the mutual information quantity I (X; Y) to become a definite quantity. Wherein, I (X; Y) can be calculated by the following formula: i (X; Y) =h (X) -H (x|y); i (X; Y) =h (Y) -H (y|x); i (X; Y) =h (X) +h (Y) -H (XY).
Fig. 3 is a schematic diagram of a collarband matrix of the sensor association network provided by the invention, as shown in fig. 3, the correlation between the sensors can be represented by a matrix form, wherein the rows and columns of the matrix respectively represent the sensor measuring points, and each element in the matrix represents the correlation degree between two sensors.
Then, each sensor is used as a single node, the correlation among the sensors is used as an edge, the nodes are connected, and the sensor association network is obtained through a spectral clustering method. Fig. 4 is a schematic diagram of a topological structure of a sensor association network provided by the invention, and as shown in fig. 4, the sensor association network has a sensor clustering relationship with clear boundaries, and the clustering relationship can reflect the functional relevance of a sensor V1. Thus, the abnormal node can be a node of the sensor association network, which falls on a bill, through the sensor abnormal point screening unit. The abnormal node is a node which may be caused by a sensor fault or a system fault in the sensor association network. The abnormal sensor in each sensor may be obtained by a sensor failure determination method based on the time domain/frequency domain characteristics.
The system provided by the embodiment of the invention obtains the abnormal node in the sensor association network based on the construction of the sensor association network, and realizes the detection of the sensor fault of the entity system.
Considering that the detection of the failure of the entity system is only realized initially after the abnormal node in the sensor association network is obtained, it is not possible to determine specifically whether the entity system has failed. Based on any one of the above embodiments, the sensor outlier screening unit further includes an outlier detection unit;
the abnormal signal detection unit is used for obtaining a fault detection result based on the transmission result of the abnormal signal corresponding to the abnormal node in the sensor association network.
Specifically, the transmission result of the abnormal signal corresponding to the abnormal node in the sensor association network can be obtained through the abnormal signal detection unit, and the transmission result can reflect whether the sensor signal exists or not. It can be understood that when the sensor fails, if the system fails, an abnormal node will also appear, so that whether the device fails when the sensor fails can be further determined by whether an abnormal signal corresponding to the abnormal node can be propagated in the sensor association network. If the abnormal signal is detected to be unable to propagate in the sensor association network, indicating that the system of the entity system 110 has not failed, the failure detection result is a sensor failure; if the abnormal signal is detected to be capable of being transmitted in the sensor association network, the system of the entity system 110 is indicated to be faulty, and the fault detection result is a system fault.
According to the system provided by the embodiment of the invention, the accurate fault detection result of the entity system is obtained through the transmission result of the abnormal signal corresponding to the abnormal node in the sensor association network.
Based on any of the above embodiments, the simulation system 120 is configured to control the entity system 110 and the simulation system 120 to operate synchronously based on a simulation control parameter, and adjust the simulation control parameter and the control rate based on the switching conditions of the entity system 110 under each normal working condition and different degradation states.
Specifically, the entity system 110 and the simulation system 120 may be controlled to run synchronously by using the simulation control parameters, the control rate is adjusted by using the switching conditions of the entity system 110 under each normal working condition and/or different degradation states, and the simulation control parameters are adjusted by using the adjusted control rate, so as to realize the control of the entity system 110 by the simulation system 120. The normal working conditions refer to working conditions of the entity system except accident working conditions and extreme working conditions. The different degradation states here refer to the degree of degradation of the various devices of the physical system. It should be noted that, the simulation control parameters or the physical control parameters herein refer to system parameters, such as control parameters of valve opening, pump rotation speed, and the like. The control rate here may reflect the control degree of the simulation control parameters under different switching conditions on each index of the entity system 110. For example, under a certain switching condition, the simulation control parameter needs to realize higher control on the response speed of the entity system 110; for another example, under certain switching conditions, the simulation control parameters need to achieve higher control over the physical system 110 in terms of corresponding accuracy. Therefore, a set of control strategies aiming at the entity system 110 can be formed through the integrated design of the simulation control parameters and the control rate according to the switching conditions of the working conditions, so as to realize the control optimization of the simulation system 120 on the entity system 110.
It will be appreciated that the ability to test simulation and analysis is limited by the inherent conditions of the test bed of the physical system 110, and is limited, and cannot fully envelope all operating conditions within the system operating regime, some high operating regime tests are costly, and some extreme and accident operating regime tests are difficult to develop. Challenges faced by the system test system include: the system multi-working condition dynamic characteristic test relates to multiple test equipment and guarantee facilities, and has the advantages of complex test flow, long debugging period and high test cost; when the system runs under a low working condition, the system is far away from the design working condition, abnormal vibration of the system is easy to induce, and the engineering test bed and the traditional measurement means are difficult to reveal the mechanism and the rule of the related process; the system limit working condition test and the accident simulation test have certain destructiveness and danger, and the test bed is required to be adaptively modified, so that the test cost is increased. Aiming at the problem, the simulation system is also used for acquiring a target simulation operation result output under a target working condition;
the target working condition comprises at least one of an extreme working condition and an accident working condition
Specifically, the simulation system 120 can only output the simulation operation result under at least one of the extreme working condition and the accident working condition or the working environment as the target policy operation result, thereby improving the simulation of the simulation system to the extreme working condition and the accident working condition, dynamically updating the safety boundary, and providing basis for fault diagnosis, prediction and safety decision of the actual system.
Based on any one of the above embodiments, the industrial virtual-real bidirectional interactive system further includes a data processing unit;
and the data processing unit is used for carrying out missing data complement on the entity operation result, the simulation operation result and the test data.
It should be noted that the data processing unit may establish a communication connection with the simulation system 120 and the entity system 110. The data processing unit is constructed by selecting a long and short time memory network in consideration of the characteristic that the data generated in a dynamic system is generally in the time sequence. Specifically, the entity operation result, the simulation operation result and the test data containing the missing data are input into the long-short-time memory network, and the missing data are subjected to data complementation through the long-short-time memory network, so that more complete data are obtained. Here, fig. 5 is a block diagram of a long-short-time memory network according to the present invention, and as shown in fig. 5, missing data of a missing part in data is obtained by inputting data into the long-short-time memory network, extracting features from a time sequence of the data by an encoding layer, and decoding the extracted features by a decoding layer. Wherein X is T Representing complete historical data with T slots, i.e., a multidimensional time series, without missing values. Xi is the d-dimensional data point of the sensor at time i. The encoder and decoder consist of T long and short timing sequences. After encoding, the bottleneck characteristic is input into a decoder to recover the missing value, and residual errors between the recovered value and the input data of the adjustment network are calculated through square errors. As with other deep learning models, training of long and short term memory networks uses a back propagation algorithm, the Adam method. The number of neurons in each layer of the long-short-time memory network needs to be set according to the input data condition, then the training is repeated for multiple times, the optimal combination is obtained, and then the long-short-time memory network is optimized, wherein the number of neurons in the middle layer is determined by a main analysis method.
According to the system provided by the embodiment of the invention, the data processing unit based on the long-short-time memory network is used for processing the real data interacted by the industrial virtual-real two-way interaction system corresponding to the dynamic system and the non-ideal data in the simulation data, so that the missing data in the data generated in the dynamic system is made up, and the industrial virtual-real two-way interaction system is excellent in tasks such as entity system state evaluation, simulation system updating, entity system fault diagnosis and the like.
Based on any of the above embodiments, fig. 6 is a second schematic structural diagram of an industrial virtual-real bidirectional interactive system according to the present invention, as shown in fig. 6, the system includes: the system comprises an entity system 110, a simulation system 120, a correction module 130, a data processing unit 140, an entity system detection module 150, a control rate optimizing module 160 and a test result output module 170 under working conditions and accident working conditions.
Specifically, the data processing unit 140 may perform data processing on test data, entity operation results, and simulation operation results.
In the industrial virtual-real bidirectional interactive system, the real control virtual can be used for collecting test data through the entity system 110 and transmitting the test data to the correction module 130, and the correction module 130 determines whether to update the simulation system 120 through switching data in the test data. When the switch data representation of the adjustment switch is updated, then the simulation system 120 is updated; when the switch data of the adjustment switch indicates that no update is performed, then no update is performed to the simulation system 120.
In addition, in the industrial virtual-real bidirectional interactive system, the entity system monitoring module 150 can perform linkage through real control and virtual control, that is, can output a simulation operation result under the same physical data through the dynamically updated simulation system 120, and perform result analysis on an entity operation result output by the entity system 110 under the same physical data, and when the analysis result is that convergence is possible, can continuously obtain the simulation operation result and the entity operation result, so as to realize monitoring of the entity system 110. When the analysis result is non-convergence, fault detection is performed on the physical system 110.
In the industrial virtual-real bidirectional interactive system, the virtual control real simulation system 110 can also be used for controlling the entity system 120 to perform control rate optimization under different working conditions so as to realize more excellent control on the entity system 110. In addition, the test result output under the extreme working condition and the accident working condition can be realized through the simulation system 120.
Based on any of the above embodiments, fig. 7 is a schematic flow chart of an industrial virtual-real bidirectional interaction method provided by the present invention, as shown in fig. 7, the method includes:
step 710, acquiring test data, wherein the test data is acquired based on an entity system in any system state;
step 720, obtaining a parameter identification result of the entity system in the current state based on the test data;
step 730, updating the simulation system based on the parameter identification result;
the system state is determined by at least one of a working condition, a life span, and a working environment of the physical system.
It should be noted that, in the industrial virtual-real bidirectional interactive system, information interaction between the systems can be realized through a communication interface between the physical system simulation system and the correction module. FIG. 8 is a functional schematic diagram of a communication interface provided by the present invention, where, as shown in FIG. 8, the communication interface may provide an interaction environment for interaction between an entity system and a simulation system as an intermediate for data exchange between the simulation system and the entity system of a controlled object. For example, control signals and process data of the entity system are collected and converted into digital signals required by the simulation system; the digital control signals sent by the simulation system can be converted into various physical signals which can be collected by a rack controller of the entity system. In addition, the simulation system may be connected in series or in parallel with the correction module. Fig. 9 is a schematic diagram of connection between a simulation system and a correction module, as shown in fig. 9, fig. a is a schematic diagram of connection between the simulation system and the correction module, and fig. B and fig. C are schematic diagrams of connection between the simulation system and the correction module. In the figure, x k Representing a system state at a kth time; u (u) k Represents x k Test data, p, represent simulation system parameters; z -1 Representing forward and backward propagation.
Specifically, the actual data output can be used as test data by performing industrial operations through the physical system in any system state. The system state herein may be determined by at least one of a certain working condition of the entity system, a life span of the entity system, and a working environment of the entity system. For example, the actual data output from the physical bench at the middle and late life spans can be used as test data. The test data herein may include numerical data, spectral data, and the like.
It should be noted that, in the actual application process of the entity system, before dynamically adjusting the simulation system, the current system state of the entity system needs to be clarified, that is, by judging whether the current system state of the entity system needs to update the simulation system, the adjusted simulation system is more in line with the entity system in the current system state. Thus, the test data may also include switching data, i.e., whether to update the simulation system corresponding to the physical system of the current system state. The switching data may be obtained by determining key operation parameters used for reflecting the working condition in the entity system, or may be determined by a value of a power value of a main pump in the entity system when the working condition changes, for example, when the power value is smaller than a preset threshold, the switching data may be turned on, that is, the simulation system of the current system state is updated; the switch data may also be determined for the life span of the physical system, for example, a "health factor" is constructed by selecting a feature that reflects the health status of the physical system, where the health factor may be normalized to (0, 1) and 0.1 is a sign of significant degradation.
According to the industrial virtual-real bidirectional interaction method provided by the embodiment of the invention, the simulation system is updated based on the test data collected by the entity system, so that the dynamic update of the simulation system is realized when the working condition, the service life period and the working environment change, the conditions of the abnormality of the entity system and the working condition, the service life period and the working environment change can be distinguished, and the system performance of the industrial virtual-real bidirectional interaction system is improved.
Based on any of the above embodiments, step 730 may be followed by:
acquiring an entity operation result and a simulation operation result which are respectively output by the entity system and the simulation system under the same physical data;
based on the entity operation result and the simulation operation result, judging the operation state of the entity system;
and continuously acquiring the entity operation result and the simulation operation result under the condition that the operation state is normal.
Based on any of the foregoing embodiments, the determining the operating state of the entity system further includes:
under the condition that the running state is abnormal, acquiring the correlation among the sensors of the entity system, taking the sensors as nodes, and constructing a sensor correlation network of the entity system by taking the correlation surrounding edges among the sensors;
based on a sensor association network, obtaining an abnormal node in the sensor association network;
and obtaining a fault detection result corresponding to the abnormal node based on a transmission result of the abnormal signal corresponding to the abnormal node in the sensor association network.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. An industrial virtual-real two-way interactive system, comprising: the system comprises an entity system, a simulation system and a correction module;
the entity system is used for collecting test data and transmitting the test data to the correction module;
the correction module is used for outputting a parameter identification result of the entity system in the current system state based on the test data and updating the simulation system based on the parameter identification result;
the test data is collected based on the physical system under any system state, wherein the system state is determined by at least one of working condition, service life and working environment of the physical system.
2. The industrial virtual-to-actual two-way interaction system according to claim 1, further comprising an entity system monitoring module, wherein the entity system monitoring module establishes communication connection with the entity system and the simulation system, respectively;
the entity system monitoring module is used for respectively outputting an entity operation result and a simulation operation result based on the entity system and the simulation system under the same entity data, and carrying out error analysis on the entity operation result and the simulation operation result to judge the operation state of the entity system;
and continuously acquiring the entity operation result and the simulation operation result under the condition that the operation state is normal.
3. The industrial virtual-real bidirectional interactive system according to claim 2, wherein the entity system monitoring module further comprises a fault detection module, the fault detection module is configured to perform fault detection on the entity system when the operation state is abnormal;
the fault detection module comprises a sensor evaluation unit, a sensor association network construction unit and a sensor abnormal point screening unit;
the sensor evaluation unit is used for acquiring the correlation among the sensors contained in the entity system;
the sensor association network construction unit is used for constructing a sensor association network of the entity system by taking the sensors as nodes and the correlation among the sensors as edges;
the sensor abnormal point screening unit is used for screening nodes of the sensor association network to obtain an abnormal sensor in the sensor association network.
4. The industrial virtual-real two-way interactive system according to claim 3, wherein said sensor outlier screening unit further comprises an outlier detection unit;
the abnormal signal detection unit is used for obtaining a fault detection result based on the transmission result of the abnormal signal corresponding to the abnormal node in the sensor association network.
5. The industrial virtual-actual bidirectional interactive system according to claim 1, wherein the simulation system is configured to control the entity system to operate synchronously with the simulation system based on a simulation control parameter, and adjust the simulation control parameter and the control rate based on a switching condition of the entity system under each normal working condition and different degradation states.
6. The industrial virtual-real two-way interactive system according to claim 5, wherein said simulation system is further configured to obtain a target simulation operation result output under a target working condition;
the target operating condition includes at least one of an extreme operating condition and an accident operating condition.
7. The industrial virtual-to-actual two-way interaction system according to any of claims 2-4, further comprising a data processing unit;
and the data processing unit is used for carrying out missing data complement on the entity operation result, the simulation operation result and the test data.
8. An industrial virtual-real two-way interaction method is characterized by comprising the following steps:
acquiring test data, wherein the test data is acquired based on an entity system in any system state;
based on the test data, obtaining a parameter identification result of the entity system in the current state;
updating a simulation system based on the parameter identification result;
the system state is determined by at least one of a working condition, a life span, and a working environment of the physical system.
9. The industrial virtual-real two-way interaction method according to claim 8, wherein updating the simulation system based on the parameter identification result comprises:
acquiring an entity operation result and a simulation operation result which are respectively output by the entity system and the simulation system under the same physical data;
based on the entity operation result and the simulation operation result, performing error analysis on the entity operation result and the simulation operation result, and judging the operation state of the entity system;
and continuously acquiring the entity operation result and the simulation operation result under the condition that the operation state is normal.
10. The industrial virtual-to-actual two-way interaction method of claim 9, wherein the determining the operational state of the physical system further comprises:
under the condition that the running state is abnormal, acquiring the correlation among the sensors of the entity system, taking the sensors as nodes, and constructing a sensor correlation network of the entity system by taking the correlation surrounding edges among the sensors;
based on a sensor association network, obtaining an abnormal node in the sensor association network;
and obtaining a fault detection result corresponding to the abnormal node based on a transmission result of the abnormal signal corresponding to the abnormal node in the sensor association network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311531199.6A CN117786941A (en) | 2023-11-16 | 2023-11-16 | Industrial virtual-real bidirectional interaction system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311531199.6A CN117786941A (en) | 2023-11-16 | 2023-11-16 | Industrial virtual-real bidirectional interaction system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117786941A true CN117786941A (en) | 2024-03-29 |
Family
ID=90382240
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311531199.6A Pending CN117786941A (en) | 2023-11-16 | 2023-11-16 | Industrial virtual-real bidirectional interaction system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117786941A (en) |
-
2023
- 2023-11-16 CN CN202311531199.6A patent/CN117786941A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112202736B (en) | Communication network anomaly classification method based on statistical learning and deep learning | |
| JP6609050B2 (en) | Anomalous fusion in temporal causal graphs | |
| CA2438903C (en) | Exception analysis for multimissions | |
| CN112987675B (en) | Method, device, computer equipment and medium for anomaly detection | |
| US7113988B2 (en) | Proactive on-line diagnostics in a manageable network | |
| JP5278310B2 (en) | Diagnostic system | |
| CN108418841A (en) | Next-generation key message infrastructure network Security Situation Awareness Systems based on AI | |
| US11675799B2 (en) | Anomaly detection system | |
| KR20180108446A (en) | System and method for management of ict infra | |
| CN104615122B (en) | A kind of industry control signal detection system and detection method | |
| JP2009053938A (en) | Equipment diagnosing system and equipment-diagnosing method on the basis of multiple model | |
| CN112101431A (en) | Electronic equipment fault diagnosis system | |
| Ntalampiras et al. | A fault diagnosis system for interdependent critical infrastructures based on HMMs | |
| KR102455332B1 (en) | Methods and devices for determining the state of a network device | |
| CN110188837A (en) | A kind of MVB network fault diagnosis method based on fuzzy neural | |
| CN108415810A (en) | Hard disk state monitoring method and device | |
| CN116070802A (en) | Intelligent monitoring operation and maintenance method and system based on data twinning | |
| CN114048546B (en) | Method for predicting residual service life of aeroengine based on graph convolution network and unsupervised domain self-adaption | |
| CN113487086B (en) | Method, device, computer equipment and medium for predicting residual service life of equipment | |
| CN108415819A (en) | Hard disk fault tracking method and device | |
| CN117526561A (en) | Digital twinning-based transformer substation equipment abnormality monitoring and early warning method and system | |
| CN116593883A (en) | Breaker body fault diagnosis method, device and equipment of intelligent high-voltage switch and storage medium | |
| CN117786941A (en) | Industrial virtual-real bidirectional interaction system and method | |
| CN110188040A (en) | A kind of software platform for software systems fault detection and health state evaluation | |
| KR20190132223A (en) | Apparatus and method for analyzing cause of network failure |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |